1. Technical Field
The invention relates to image creating methods, particularly to an image creating method which corrects an image formed by the TOF principle.
2. Related Art
A depth camera can be used to control a computer through a gesture. Moreover, a depth camera can be further used to control a TV game through a body motion. This makes human-machine interaction more intuitive.
Such human-machine interaction needs a depth camera which can store a three-dimensional image into a two-dimensional format. A depth camera can measure a Z-axis distance between every shot point and the camera so that it can record three-dimensional image data.
A common method for measuring the Z-axis distance is to use the principle of time of flight (TOF). Simply speaking, a time period from a light beam emitted by a light source to be reflected by a shot point to come back to the origin can be used to calculate the Z-axis distance.
The Z-axis distance measured by the TOF principle is the distance between the lens and each shot point. Please refer to FIG. 1, distances between shot points A, B and C on a plane and the lens 100 are D1, D2 and D3, respectively. Distances D1, D2 and D3 are obtained by the formula: d=(phase-offset)R/2π. But in practice, cos θ1 and cos θ2 will have an error because of an optical error of the light source. As a result, the image cannot be created on a plane. This is a primary drawback of the TOF principle.